This project is a continuation of a long-term research effort directed toward describing and modeling the functions of cell groups within the superior olivary complex of the cat. The proposed work is a collaborative effort involving a sensory neurophysiologist, C. Tsuchitani of the University of Texas Health Science Center at Houston (UTHSC-H), and a communication- signal processing theorist, D.H. Johnson of Rice University. The objective of this project is to determine the manner in which auditory information is encoded in the responses of lower auditory neurons by utilizing statistical communications theory, i.e., the theory of point processes, as the mathematical framework of the representation of single unit discharge patterns. Binaural hearing provides significant advantages over monaural hearing in perceptual tasks such as sound detection, localization of sound sources and speech intelligibility in a noisy background. The lateral superior olive (LSO) is believed to be of primary importance in binaural processing of high (greater than 1 kHz) frequency information. The proposed research is concentrating on the neural network which gives rise to the responses of the LSO neurons to stimulation of the two ears. The single unit discharges of LSO units and units presumed to provide inputs to the LSO (i.e., the primary-like units of the anteroventral cochlear nucleus and of the medial nucleus of the trapezoid body) are recorded extracellularly. Histological methods will be used to confirm unit location with the aid of deposits made with the recording electrode. Monaural stimuli are used to elicit responses that are used to establish the form of the statistical model of unit discharges. LSO unit discharges to binaural stimuli are providing measures of the statistics of the inhibitory response and will be used to develop the model of LSO unit binaural responses. Once the models of the LSO and its inputs have been established, the LSO network model will be developed and its consistency checked by comparing input and output quantities. Models of binaural localization will be studied to develop estimates of the limits of localization acuity imposed by physical constraints. These limits will provide a baseline measure of performance which can be compared with the limits placed on acuity by the model of LSO unit processing of binaural information. Data collection and initial data analysis are carried out at the UTHSC-H facility under the direction of Dr. C. Tsuchitani. Dr. D.H. Johnson directs the statistical modeling of the single unit discharges at Rice University.